Hydraulic water permeability and transepithelial voltage in the isolated perfused rabbit cortical collecting tubule following acute unilateral ureteral obstruction.

Ureteral obstruction affects the kidney's ability to conserve water and sodium. Using the isolated perfused tubule technique, we studied cortical collecting tubules (CCT) taken from rabbits subjected to a sham operation or to 4 h of unilateral ureteral obstruction (UUO). Tubules were perfused in the presence of an osmotic gradient directed to promote water movement from lumen to bath, and volume flux (Jv), hydraulic water permeability (Lp), and transepithelial voltage (V1) were determined. In tubules from sham-operated and UUO animals, basal (before exposure to vasopressin) J, and Lp were not different from zero. After addition of 200 microU . ml-1 of arginine vasopressin (aVP) to the bath, Jv and Lp increased to 1.64 +/- 0.23 nl . mm-1 . min-1 and 127.9 +/- 19.8 cm . s-1 . atm-1 x 10(7), respectively, in tubules from sham-operated animals, but not only 0.27 +/- 0.09 nl . mm-1 . min-1 an 18.8 +/- 6.2 cm . s-1 . atm-1 . 10(7) in tubules from UUO animals. Pretreatment with desoxycorticosterone acetate (DOCA) or indomethacin in vivo did not prevent the blunted vasopressin response seen in tubules taken from UUO animals. The Jv and Lp responses to the cyclic AMP (cAMP) analogue, 8-Br-cAMP, were also diminished in tubules taken from UUO animals compared with shams. V1, measured during the basal period, was diminished in tubules from UUO kidneys (-5.0 +/- 2.1 mV) compared with shams (-21.9 +/- 4.1 mV), and pretreatment with DOCA did no prevent the effects of UUO on V1. In contrast, tubules taken from animals that received indomethacin prior to UUO developed voltages not different from voltages in tubules taken from sham-operated animals (-17.3 +/- 1.7 mV). We conclude that, although CCT from UUO animals can maintain osmotic gradients, their ability to respond to vasopressin by increasing Lp is impaired by an intrinsic defect located at a step beyond the generation of cAMP, and that prostaglandin inhibition or DOCA pretreatment do not reverse the decreased responsiveness of Lp to aVP. UUO also diminished V1, and this abnormality was prevented by previous treatment with indomethacin, suggesting that prostaglandins may mediate the effect of UUO on V1.

Effects of acute unilateral renal denervation in the rat.

Studies were undertaken to characterize the renal responses to acute unilateral renal denervation and the mechanisms involved in these responses. Denervation was produced in anesthetized nondiuretic rats by application of phenol to the left renal artery. Studies were also performed in sham-denervated nondiuretic rats. Whole kidney and individual nephron studies were performed before and after denervation or sham denervation. Denervation increased urine volume from the left kidney to about twice its control value (P less than 0.001) and increased urinary sodium excretion from 332 neq min minus -1 to 1,887 neq min minus -1 (P less than 0.001). Glomerular filtration rate (GFR) and renal plasma flow (RPF) remained unchanged in both kidneys after the procedure. The innervated right kidney showed no changes in urine volume or in sodium excretion. After denervation, late proximal ratio of tubular fluid inulin concentration to that of plasma [(F/P)In] decreased from 2.23 to 1.50 (P less than 0.001) while single nephron GFR remained unchanged. Absolute reabsorption decreased from 16.5 to 9.9 n. min minus -1 (P less than 0.001). (F/P)In ratios were also decreased in early distal (from 6.21 to 3.18, P less 0.001) and late distal convolutions (from 16.41 to 8.33, P less than 0.001) during the experimental period. (F/P)Na ratios remained unchanged in the early distal convolutions, but increased from 0.18 to 0.38 (P less than 0.01) in late distal convolutions after denervation. Absolute Na reabsorption after denervation increased in the loop of Henle, distal convolution, and collecting ducts. Any changes in intrarenal hydrostatic pressures after denervation were always small. There were no changes in GFR, RPF, urine volume, urinary sodium excretion, or late proximal (F/P)In after sham denervation. We conclude that the diuresis and natriuresis seen after acute renal denervation were caused by a marked depression of sodium and water reabsorption in the proximal tubule with partial compensation in more distal nephron segments. These responses appeared to be unrelated to systemic or intrarenal hemodynamic changes. The results demonstrate an effect of the renal nerves on proximal tubular function.

Effects of ouabain on fluid transport and electrical properties of Necturus gallbladder. Evidence in favor of a neutral basolateral sodium transport mechanism.

Net fluid transport (Jv) and electrical properties of the cell membranes and paracellular pathway of Necturus gallbladder epithelium were studied before and after the addition of ouabain (10(-4) M) to the serosal bathing medium. The glycoside inhibited Jv by 70% in 15 min and by 100% in 30 min. In contrast, the potentials across both cell membranes did not decrease significantly until 20 min of exposure to ouabain. At 30 min, the basolateral membrane potential (Vcs) fell only by ca 7 mV. If basolateral Na transport were electrogenic, with a coupling ratio (Na:K) of 3:2, the reductions of Vcs at 15 and 30 min should be 12--15 and 17--21 mV, respectively. Thus, we conclude that the mechanism of Na transport from the cells to the serosal bathing solution is not electrogenic under normal transport conditions. The slow depolarization observed in ouabain is caused by a fall of intracellular K concentration, and by a decrease in basolateral cell membrane K permeability. Prolonged exposure to ouabain results also in an increase in paracellular K selectivity, with no change of P Na/P Cl.

P-glycoprotein functions and substrates: possible roles of MDR1 gene in the kidney.

There is a renewed attention on the multidrug resistance genes and their products, P-glycoproteins, since recent molecular and functional studies revealed unexpected functions in normal tissues. There are two types of human P-glycoprotein: Type I, encoded by the MDR1 gene, present in excretory organs and in non-polarized cells; and Type II, encoded by MDR2, present in the canalicular membrane of hepatocytes. MDR1 Pgp transports xenobiotics, peptides, steroids, and phospholipids, and is also a regulator of swelling-activated chloride channels. MDR2 Pgp is exclusively a phosphatidylcholine translocase. In the kidney, the MDR1 gene and protein are expressed in mesangial, proximal tubule, thick loop of Henle, and collecting duct cells. In mesangial and proximal tubule cells Pgp transports xenobiotics. Concomitant exposure of kidney cells to two Pgp substrates results in increased cell toxicity. Extracts from supernatants of mesangial cell cultures inhibit Pgp-mediated transport, suggesting that a mesangial-cell metabolite could be a substrate of Pgp. Active vitamin D3 and platelet activating factor inhibit Pgp transport and are possible endogenous substrates in proximal tubule and mesangial cells, respectively. Pgp could be also a regulator of swelling-activated chloride channels present in the kidney.

Inhibition of tubule-cell proliferation to prevent cyst formation: a new avenue to treat ADPKD?

Tubule-cell hyperproliferation precedes cyst development in autosomal dominant polycystic kidney disease (ADPKD). Parker et al. report that insulin-like growth factor-1 stimulates ADPKD cell proliferation by activating Ras and Raf; inhibition of Ras or Raf abolished this effect. Inhibiting tubule-cell proliferation could halt cyst formation and prolong survival of functional tubules, offering a new treatment for ADPKD patients.

Evidence of angiogenesis and microvascular regression in autosomal-dominant polycystic kidney disease kidneys: a corrosion cast study.

Autosomal-dominant polycystic kidney disease (ADPKD) accounts for about 10% of all cases of chronic renal failure requiring dialysis. The disease is characterized by proliferation of renal epithelial cells and formation of cysts that expand over years and replace the normal parenchyma of the kidney. As the cysts grow, the volume of the kidney can increase by more than 10-fold, implying that remodeling and expansion of the vasculature must occur to provide oxygenation and nutrition to the cyst cells. Our previous studies support the notion that there is angiogenesis in ADPKD. We report here results from resin casting of ADPKD kidneys vasculature. In this study, the corrosion-casting method was used in conjunction with scanning electron microscopy to study the vascular architecture and the evidence for angiogenesis in ADPKD kidneys. We found a well-defined vascular network around the cysts forming a 'vascular capsule' somewhat similar to that described in avascular leiomyomata. We also found that the normal vascular architecture is lost and replaced by an assortment of capillaries of larger size than those in the normal kidney, mixed with flattened and spiral arterioles, damaged glomeruli, and atresic venules, indicative of regression of the microvasculature. In the same areas, there was capillary sprouting, considered the hallmark of angiogenesis. The present study documents regression changes of the vasculature and confirms the existence of angiogenesis in ADPKD kidneys, and suggests the use of inhibitors of angiogenesis as a possible avenue for the treatment of the disease.

Electrophysiological identification of cell types in cortical collecting duct monolayers.

Double-barrel microelectrodes were used to determine membrane voltages and the intracellular pH (pHi) in primary cultures of cortical collecting duct cells (CCD) grown in the absence of aldosterone. Electrophysiologically, two main cell types were identified. In cell type 1, the apical membrane voltage (Va) was -60 +/- 5 mV. The fractional resistance of the apical membrane (fRa) was 0.40 +/- 0.03, and pHi was 7.21 +/- 0.04. Exposure to 50 mM K+ on the apical side depolarized Va by 21 +/- 4 mV. When Cl- was replaced by cyclamate two types of responses were observed: (a) depolarization of Va by 26 +/- 3 mV while pHi remained unchanged, and (b) no change in Va. In cell type 2, Va was -36 +/- 5 mV, fRa was 0.91 +/- 0.03 and increasing apical [K+] from 5 to 50 mM did not change Va. Two subpopulations were distinguished by the response of pHi to lowering apical [Cl-]. In one of them pHi increased from 6.99 +/- 0.05 to 7.11 +/- 0.07. In the other, pHi was significantly decreased from 7.16 +/- 0.08 to 7.03 +/- 0.07. These results are compatible with the conclusion that about 50% of the impaled cells type 2 have a Cl-/HCO-3 exchanger at the apical membrane. In summary, two different cell types can be identified electrophysiologically in CCD monolayers. Cell type 1 has the electrical characteristics of principal cells. Cell type 2 resembles the intercalated cells. The cell alkalinization observed in approximately 50% of the cells type 2 in response to Cl- removal suggests the presence of an apical Cl-/HCO-3 exchanger. Thus, these cells should be the bicarbonate-secreting cells. The remaining cells should correspond to the acid-secreting cells.

Characterization of peanut-lectin (+) cells derived from the RCCT-28A cell line.

Cells from RCCT-28A cell line, which exhibited the highest peanut-lectin binding capacity [RCCT-28A(P+)] were separated and cultured attempting to obtain a population of HCO3- secreting cells. The mechanisms of apical solution acid or base extrusion were studied in confluent monolayers. In RCCT-28A (P+) cells net H+ flux (JH+, nmol.min-1.cm-2) was significantly lower (JH+ = 13 +/- 3) than that observed in the RCCT-28A cells (21 +/- 2). N-ethylmaleimide, Bafilomycin-A, omeprazole and Schering 28080 decreased JH+. Cl- removal from the basolateral side, in the presence or absence of SCH in the apical side, decreased JH+. Removal of apical Cl- abolished apical extrusion of base-equivalents. Incubation in a high bicarbonate solution results in base-equivalent flux to the apical side (JH+ = -7.0 +/- 1.0). Preincubation in a hyperosmotic medium (mannitol addition) at pH 7.5 decreased JH+ to -3.0 +/- 1.0. We conclude that, functionally, RCCT-28A(P+) cells are only quantitatively different from RCCT-28A cells. Acid secretion by these cells can be modulated by alkaline and by hyperosmolar preincubation and is most likely mediated by coexistent of apical H(+)-ATPase and H+,K(+)-ATPase. The main mechanism of net OH- secretion appears to be Cl-/HCO3- exchange at the apical side.

Characterization of acid-base transport mechanisms in the kidney cell line RCCT-28A.

RCCT-28A cells, a continuous cell line of rabbit cortical collecting tubule origin, have been found to exhibit apical peanut-lectin binding, basal band-3 immunostaining and a transepithelial electrical resistance of 246 +/- 37 omega cm2. For the studies reported, confluent monolayers of RCCT-28A cells were grown on permeable wells and incubated in a control solution or in alkaline solutions by lowering PCO2. Equivalent H+ fluxes (JH+) into the apical solution (nmol.min-1.cm2) were measured in the absence of drugs and in the presence of: amiloride (A, 10(-3) M), N-ethylmaleimide (NEM, 5 mM) and omeprazole (OM, 100 microM) in the apical solution. After preincubation in control solutions JH+ was 21 +/- 2 while A had no effect. Addition of NEM diminished JH+ to 12 +/- 2 (P < 0.005), and OM diminished JH+ to 2 +/- 2 (P < 0.001 vs. control). Monolayers incubated at low PCO2 had a basal JH+ of 11 +/- 5. No effect on JH+ could be demonstrated under these conditions by addition of NEM or OM. Removal of K+ from the apical solution diminished apical acidification by 60%. The inhibitor of H+,K(+)-ATPase Schering 28080 (SCH) was tested at different concentrations and an inhibitory effect was demonstrated (JH+ -2 +/- 1 vs. 18 +/- 1, SCH vs. control, respectively). Probenecid and bafilomycin-A also decreased apical acidification and an apical base-equivalent extrusion was apparent under the inhibitors effect. JH+ was abolished by removal of Cl- from the basolateral solution.(ABSTRACT TRUNCATED AT 250 WORDS)

Cell membranes and paracellular resistances in isolated renal proximal tubules from rabbit and Ambystoma.

Transepithelial specific resistance (Re) was measured in isolated and perfused rabbit proximal convoluted tubules by cable analysis and intracellular micro-electrode techniques were used to calculate the electrical resistances of the cell membranes and of the paracellular pathway. Re was 16 +/- 2 omega cm2 and the space constant was 130 +/- 14 micron, n = 29. Re was significantly increased by a decrease in temperature from 37 to 10 degrees C, and was practically abolished by nominal removal of Ca2+ from the bathing solution (to 2.0 +/- 0.3 omega cm2, P less than 0.001, n = 6). The apparent ratio of cell membrane resistances (luminal to basolateral) was 3.1 +/- 0.3. The control values of apical and basolateral membrane resistances (Ra and Rb) were calculated from the values of (1) Re, (2) the apparent ratio of cell membrane resistances, and (3) the effects of addition of either Ba2+ (1 mM) to the bath solution or glucose (8 mM) to the perfusate on basolateral and apical membrane voltages (assuming that the initial effects of Ba2+ and glucose are restricted to the ipsilateral membrane). Control values of Ra (omega cm2 of epithelium) were 249 +/- 68 (Ba2+ method) and 227 +/- 42 (glucose method). Values of Rb were 70 +/- 11; and 66 +/- 12 respectively. The low paracellular resistance values obtained with the Ba2+ and glucose methods, respectively, 17 +/- 5 and 15 +/- 1 omega cm2, explain the low transepithelial resistance. The use of the Ba2+ and glucose methods provides alternatives to cell cable determinations for the calculation of cell membrane resistances. Cell membrane and shunt resistances measured by the same methods in isolated perfused Ambystoma tigrinum proximal tubules (in omega cm2 of epithelium) were: Ra, 2650 +/- 180 (glucose method) and 2368 +/- 350 (Ba2+ method). Values of Rb were 665 +/- 99 (glucose method) and 701 +/- 124 (Ba2+ method). The paracellular resistance values were 58 +/- 11 (glucose method) and 84 +/- 12 (Ba2+ method). These results are in good agreement with previously reported values obtained by intracellular cable analysis (Maunsbach & Boulpaep, 1984).

Electrical properties of the basolateral membrane of the straight portion of the rabbit proximal renal tubule.

1. Cell membrane potentials were measured with intracellular 3 M-CKl microelectrodes in isolated, perfused segments of the straight portion of the rabbit proximal tubule. 2. Under in vitro conditions simulating the in vivo situation, the transepithelial potential difference was about 1.6 mV, lumen-positive, and the basolateral membrane potential was 61 mV, cell negative. 3. Isomolar single ion substitutions in the bath (K+ for Na+, isethionate for Cl-, N-methyl-D-glucamine (NMDG+) for Na+, and Cl- for HCO3-) resulted in significant basolateral membrane potential changes only when [K+] was increased and [HCO3-] was reduced; in both cases the basolateral membrane depolarized. Cl- and Na+ substitutions with large monovalent ions did not change basolateral membrane potential. 4. Transepithelial potential changes in substitution experiments suggest that, at the paracellular pathway, PK greater than PNa greater than PNMDG, and PCl greater than Pisethionate. 5. It is concluded that the basolateral membrane of these cells is mainly K+-conductive and that electrodiffusional PNa and PCl are undetectable by this technique. 6. Addition of 1 mM-Ba2+ to the bath reduced basolateral membrane electro-diffusional PK, as evidenced by depolarization and by a reduction of the magnitude of the change in membrane potential produced by increasing bath [K+]. 7. The depolarization produced by lowering bath [HCO3-] appears to result from a reduction of electrodiffusional PK, since it is blocked by Ba2+. There is no need to postulate a conductive pathway for HCO3- or a related species.

Effect of catecholamines on fluid reabsorption by the isolated proximal convoluted tubule.

The effect of L-norepinephrine and isoproterenol on fluid transport was studied in superficial convoluted segments and straight portions of the rabbit proximal tubule (PCT and PR, respectively) by the technique of microperfusion in vitro. In PCT, L-norepinephrine (10(-6) M) added to the bath stimulated reversibly fluid reabsorption (Jv) by about 67%. In the presence of propranolol (10(-4) M) in the bath, L-norepinephrine caused a small (about 19%) but significant decrease of Jv. Phentolamine and isoproterenol added simultaneously to the bath also increased fluid reabsorption, an effect that was abolished by propranolol. Norepinephrine had no effect on Jv when added to the perfusate in PCT. No change of Jv was observed after addition of norepinephrine to the bath in PR. The effects on fluid reabsorption rate observed in PCT are consistent with a physiologic role of the sympathetic nervous system in the modulation of PCT fluid transport.

Effects of sodium orthovanadate on whole kidney and single nephron function.

The renal effects of sodium vanadate (Na3VO4), an inhibitor of sodium-potassium-ATPase recently shown to be a potent diuretic, were studied by using clearance and micropuncture techniques in nondiuretic anesthetized rats. Administration of 1.0 mumole of sodium vanadate (high dose) increased urine flow rate (V) from 9.8 +/- 1.4 to 17.5 +/- 4.0 microliter/min (mean +/- SEM, P < 0.025), UNaF from 1.73 +/- 0.36 to 3.05 +/- 0.65 microEq/min (P < 0.025), and FENa from 0.67 +/- 0.15 to 1.24 +/- 0.28% (P < 0.025)., No significant changes in GFR or RPF were observed. Late proximal tubular-fluid-to-plasma (F/P) inulin decreased from 2.28 +/- 0.19 to a minimum value of 1.38 +/- 0.06 (P < 0.025). Absolute water reabsorption decreased from 15.8 +/- 3.5 to 6.5 +/- 1.7 nl/min (P < 0.025) and fractional water reabsorption from 52.0 +/- 4.4 to 26.5 +/- 4.1% (P < 0.025). The injection of 0.5 mumole of sodium vanadate (low dose) resulted in no significant changes in V. Late proximal F/P inulin decreased, however, from 2.37 +/- 0.14 to a minimum value of 1.59 +/- 0.12 (P < 0.025). SNGFR remained unchanged, as did GFR and RPF. UNaV increased from 1.41 +/- 0.35 to 2.25 +/- 0.35 microEq/min (P < 0.025), and FENa rose from 0.64 +/- 0.16 to 0.91 +/- 0.15% (P < 0.025). The decrease in F/P inulin was observed in all but one animal, even in the absence of a diuretic response. The amount of fluid remaining in the lumen of the late proximal tubule was virtually the same in both low- and high-dose animals (18.9 +/- 3.0 and 19.5 +/- 3.4 nl/min, respectively). We conclude that sodium vanadate causes a decrease in superficial proximal tubule fluid and salt reabsorption. Inasmuch as the low dose does not result necessarily in a diuretic response, an increase in fluid reabsorption distal to the late proximal tubule must take place.

Electrophysiological studies on primary cultures of proximal tubule cells.

Primary confluent monolayers were grown from proximal tubule fragments of rabbit kidneys. The fragments were obtained by gradient centrifugation and seeded on an ad hoc dish whose bottom was a permeable and transparent collagen membrane. The culture medium was a mixture of 50% Ham's F-12 and 50% Dulbecco's modified Eagle's medium supplemented with insulin, transferrin, ethanolamine, sodium selenite, and amino acids. The monolayers were studied at 6-14 days after seeding. Transmission electron microscopy revealed cuboidal cells 8.5-10.5 microns high, with a 1.5 to 2.5-microns apical brush border, abundant mitochondria, vacuoles, lysosomes, and irregular basal interdigitating processes. Cyclic AMP synthesis was stimulated by parathyroid hormone and was insensitive to vasopressin and isoproterenol. Electrophysiological studies performed with the same physiological salt solution on both sides revealed a transepithelial voltage of -2.6 +/- 0.6 mV (n = 10) and a basolateral membrane voltage of -51.0 +/- 4.5 mV (n = 13), both referred to the basal solution. The transepithelial electrical resistance was 7 +/- 2 omega X cm2. The apical membrane depolarized on addition of glucose to the apical side and hyperpolarized on removal of glucose. Changes in apical membrane voltage on addition of varying glucose concentrations (at [Na] = 135 mM, 37 degrees C) demonstrate the presence of a glucose transport system with an apparent Km of 3.54 +/- 0.54 and a Vmax of 7.2 +/- 0.4 mV. Thus this preparation exhibits morphological and electrophysiological characteristics of proximal tubule cells; these studies demonstrate the feasibility of the use of intracellular microelectrode techniques to study the transport properties of cultured epithelia.

Expression and function of P-glycoprotein in human mesangial cells.

P-glycoprotein (PGP), responsible for multidrug resistance (MDR) in cancer cells, is normally expressed in kidney proximal tubules and mesangium. PGP expression and function were studied in human mesangial cell cultures. MDR1 gene expression was demonstrated by reverse transcription-polymerase chain reaction. PGP expression was determined using MRK16 monoclonal antibody and its function was assessed by the efflux of rhodamine-123 (R123). R123 efflux had a half time of 25 +/- 5 s. Efflux was inhibited by cyclosporin A (10 microM), verapamil (10 microM), and vinblastine (100 microM) with half times of 380, 535, and 312 s, respectively. Incubation with MDR1-antisense oligonucleotide decreased R123 efflux (half time = 304 s). Verapamil, cyclosporin A, and PSC-833 augmented the cytotoxicity of Adriamycin by reducing the 50% maximal growth-inhibitory dose from 730 nM to 130, 110, and 90 nM, respectively. We conclude that human mesangial cells express MDR1 and demonstrate xenobiotic transport inhibitable by several known PGP substrates. Concomitant exposure of mesangial cells to PGP-transported drugs causes intracellular accumulation of toxic PGP substrates and ultimately damages the mesangial cells.

Electrophysiological studies of primary cultures of rabbit distal tubule cells.

Primary monolayers grown from F1 band of a Percoll gradient centrifugation ("distal" monolayers) were studied, after confluency, 6-14 days after seeding. Transmission and scanning electron microscopy revealed that two cell types, resembling principal cells of the rabbit cortical collecting tubule (CCT) and intercalated cells of either CCT or connecting tubule, constitute approximately 96% of the monolayer. About two-thirds of the intercalated cells fluoresced when treated with fluorescent peanut lectin. Indirect specific immunocytofluorescent staining revealed fluorescence in 96% of the cells, confirming that the monolayers were derived from CCT or connecting tubule cells. Exposure of monolayers to vasopressin or isoproterenol increases adenosine 3',5'-cyclic monophosphate (cAMP) content in the cells and bathing medium, whereas parathyroid hormone was ineffective. Electrophysiological studies revealed a transepithelial voltage (VT) of -11 +/- 2 mV, and basolateral membrane voltage (Vb) of -77 +/- 5 mV (n = 20). The transepithelial electrical resistance (RT) was 1,870 +/- 250 omega X cm2 (n = 13). In three out of six monolayers, amiloride (10(-5) M) applied to the apical side produced an increase in apical membrane voltage (Va) from -71 +/- 1 to -89 +/- 9 mV) and a decrease in VT (from -10 +/- 1 to -2 +/- 1 mV). The RT did not change during amiloride exposure. Exposure of the apical membrane to 140 mM K+-depolarized Va from -67 +/- 7 to -39 +/- 11 mV (P less than 0.002) and hyperpolarized VT from -7 +/- 2 to -15 +/- 3 mV (P less than 0.005). Exposure to high K+ from the basolateral side depolarized Vb from -76 +/- 11 to -43 +/- 10 mV (P less than 0.001) and depolarized VT from -9 +/- to 8 +/- 5 mV (P less than 0.001). This preparation is suitable to study basic aspects of epithelial transport by electrophysiological methods and other techniques. The findings are consistent with several of the known properties of cortical collecting tubules from rabbits studied by the isolated perfused tubule technique.

Expression and function of P-glycoprotein in a mouse kidney cell line.

P-glycoprotein (PGP), a transporter conferring multidrug resistance to cancer cells, is expressed in the kidney. C219 monoclonal antibody binding revealed PGP in proximal tubules and mesangium of mouse kidneys. A cell line (TKPTS) expressing PGP was developed from proximal tubules of the 8Tg(SV40E)Bri7 mouse. Northern blot analysis demonstrated a 5.0-kb message identified as mdr1 by ribonuclease protection assay. Cyclosporin A (CSA) at 0.15 and 10 microM increased cellular accumulation of verapamil (VRP) by 32 and 121%, respectively (P < 0.001). VRP at 5 microM increased steady-state cellular accumulation of CSA by 46% (P = 0.02). Basal-to-apical transport of the PGP substrate vinblastine was inhibited by VRP. Rhodamine-123 (R-123) influx was rapid and independent of PGP. R-123 efflux was inhibited by VRP and CSA. Inhibition of PGP transport by VRP, CSA, and PSC-833 decreased the 50% effective dose of adriamycin. The concomitant administration of VRP and CSA was not deleterious and coincided with preferential accumulation of VRP over CSA. Inhibition of PGP-mediated transport is demonstrated as a mechanism of renal cell toxicity.

Xenobiotic transport differences in mouse mesangial cell clones expressing mdr1 and mdr3.

P-glycoprotein (PGP), which confers multidrug resistance to cancer cells, is expressed in mouse kidney proximal tubule and mesangium. We report on the expression of PGP and its xenobiotic transport function in mesangial cells. Studies were performed in a mouse mesangial cell line (TKGM) and two cell clones. Ribonuclease protection assay and Western blot analysis demonstrated that TKGM cells expressed mdr1 and mdr3, the isoforms responsible for multidrug resistance. TKGM-F12 cells coexpressed mdr1 and mdr3 whereas TKGM-G2 cells expressed only mdr1. The drug transport function, measured by rhodamine 123 (R-123) efflux, was smaller in TKGM-F12 than in TKGM-G2 cells. The PGP substrates adriamycin, cyclosporin A, vinblastine, and verapamil inhibited R-123 transport in TKGM and TKGM-G2 cells. In the cells studied, PGP conferred some resistance to adriamycin; concomitant exposure to adriamycin with another PGP substrate impaired cell growth. The differential expression of mdr1 and mdr3 in mouse mesangial cell clones, the ability of mdr1 PGP to transport R-123, and the impairment of PGP-mediated transport in TKGM-F12 cells, coexpressing mdr1 and mdr3 products, are demonstrated. PGP may play a physiological role in mesangial cells.

Secretion of platelet-activating factor is mediated by MDR1 P-glycoprotein in cultured human mesangial cells.

MDR1 P-glycoprotein (Pgp), the product of the MDR1 gene involved in multidrug resistance in cancer cells, is also expressed in normal tissues. In the human kidney, it is localized in the mesangium, the proximal tubule, the thick ascending limb of Henle's loop, and the collecting duct. Pgp actively transports lipophilic xenobiotics, peptides, steroids, and lipids, and perhaps endogenous substrates. It has been shown previously that human mesangial cells in culture express active Pgp and that the expression of Pgp can be down-regulated by exposure to antisense oligonucleotides. Mesangial cells do not express multidrug resistance-related protein (MRP). Experiments were performed to determine whether 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine (generically platelet-activating factor, PAF) is a substrate of Pgp in human mesangial cells in culture. This study found: (1) PAF C-16 and analogs inhibited Pgp-mediated efflux of rhodamine 123 by 59 to 88% in multidrug-resistant KBV-1 cells and by 85 to 97% in cultured human mesangial cells. (2) In mesangial cells stimulated with A23187, the secretion of endogenously produced PAF was inhibited by >80% by the Pgp blockers verapamil, cyclosporin A, PSC-833, vinblastine, and adriamycin. (3) Preincubation with MDR1 antisense oligonucleotides also blocked PAF secretion by human mesangial cells. PAF analogs do not modify the transport of MRP substrates in MCF-7/VP cells expressing MRP but not Pgp. These results indicate that PAF is an endogenous substrate of Pgp in human mesangial cells. Inhibition of Pgp transport may be useful in reducing glomerular damage occurring in pathologic conditions where PAF secretion is elevated.

Effect of dopamine on phosphate reabsorption in isolated perfused rabbit proximal tubules.

The effect of dopamine (10(-6) M) on PO4 reabsorption was studied in isolated and perfused proximal convoluted and straight tubules of the rabbit. In convoluted tubules, fluid reabsorption (JV), PO4 lumen-to-bath (JPO4lb), and net (net JPO4) fluxes did not change during dopamine exposure. In S3 segments bathed in rabbit serum, JV diminished significantly during dopamine exposure. JPO4 decreased by 20% (p less than 0.005) and net JPO4 decreased by 27% (p less than 0.02). The effect of dopamine was reversible upon removal of the drug.